AI Paper Atlas

Research map / Adversarial robustness and security / Attacks on text and graphs

On the Robustness of Semantic Segmentation Models to Adversarial Attacks

Anurag Arnab, Ondrej Miksik, Philip H.S. Torr

CVPR 2018 · Adversarial robustness and security › Attacks on text and graphs · approach: Adversarial perturbations beyond images · 330 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Use a ResNet backbone instead of VGGFCN8s and Deeplab v2 ASPP · Pascal VOCadversarial robustnessImproved
Use Deeplab v2 with Multiscale ASPP instead of its single-scale variant (ablation)Deeplab v2 · Pascal VOC and Cityscapesadversarial robustnessImproved
Add the Dilated-Net Context module (ablation)Dilated-Net Front-end · Cityscapes and Pascal VOCadversarial robustnessHurt
Use a lightweight residual architecture rather than a chain-like VGG-based networkSemantic segmentation models · Pascal VOC and Cityscapesadversarial robustnessImproved
Process inputs at multiple scales (ablation)Deeplab v2 · Pascal VOCIoU ratio under adversarial attacksImproved
Generate perturbations at one scale and evaluate them at anotherDeeplab v2 and other segmentation networks · Pascal VOCattack effectivenessHurt
Add mean-field CRF inference (ablation)FCN8s · Pascal VOC, untargeted attacksadversarial robustnessImproved
Add mean-field CRF inference (ablation)FCN8s · Pascal VOC, targeted attacksadversarial robustnessNo clear effect
Generate black-box perturbations from the FCN8s unary output rather than CRF-RNN outputCRF-RNN · Pascal VOCattack effectivenessImproved
Increase CRF pairwise-potential bandwidth to produce more smoothing (ablation)CRF-RNN · Pascal VOCadversarial robustnessNo clear effect
Use single-step FGSM on Cityscapes rather than Pascal VOCSegmentation models · Cityscapes versus Pascal VOCIoU ratio under FGSM: 2.5% versus 27.9% for PSPNet; 2.8% versus 12.2% for Dilated ContextHurt
Use single-step FGSM rather than targeted Iterative FGSM ll at high perturbation levelsSegmentation models · Cityscapesattack effectivenessImproved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy